An unexpected consequence of the surgical management of some patients with heart failure is an improvement in cardiac function following mechanical unloading. Early clinical evidence in patients with dilated cardiomyopathy suggests that this improvement may even be sustained after the removal of the left ventricular assist device (LVAD). The mechanisms for this phenomenon are not understood. Because a small piece of myocardium is taken out during placement of the LVAD, and because the same heart can be studied at the time of transplantation, we have been able to obtain preliminary evidence that left ventricular unloading leads to a reprogramming of genes for metabolic pathways of energy production in the heart which is consistent with a fetal pattern of gene expression. We suggest that this phenomenon may become a new paradigm in the assessment and treatment of advanced heart failure. Specific Aim 1 is to identify and quantitate transcripts for proteins of energy substrate metabolism and to compare their levels of expression in normal, failing, and mechanically unloaded human heart. We will also quantitate transcripts of cardiac specific switches and test whether isoform switches correlate with changes in ventricular function. Specific Aim 2 is to measure changes in transcription, expression and activities of enzymes of energy substrate metabolism in a surrogate model of left ventricular unloading. Shifts in substrate metabolism will be measured in the working heart. A full understanding of gene switching will provide insight into mechanisms underlying functional recovery of the heart. Furthermore, it is expected to yield a molecular marker predicting the safe removal of the pump device. If the hypothesis is correct, activation of fetal genes by unloading may also become an alternative approach to gene therapy in the treatment of heart failure.